Function categorical_crossentropy

tensorflow/python/keras/backend.py:4322–4373

Categorical crossentropy between an output tensor and a target tensor. Arguments: target: A tensor of the same shape as `output`. output: A tensor resulting from a softmax (unless `from_logits` is True, in which case `output` is expected to be the logits). fr

(target, output, from_logits=False, axis=-1)

@keras_export('keras.backend.categorical_crossentropy')
def categorical_crossentropy(target, output, from_logits=False, axis=-1):
  """Categorical crossentropy between an output tensor and a target tensor.

  Arguments:
      target: A tensor of the same shape as `output`.
      output: A tensor resulting from a softmax
          (unless `from_logits` is True, in which
          case `output` is expected to be the logits).
      from_logits: Boolean, whether `output` is the
          result of a softmax, or is a tensor of logits.
      axis: Int specifying the channels axis. `axis=-1` corresponds to data
          format `channels_last', and `axis=1` corresponds to data format
          `channels_first`.

  Returns:
      Output tensor.

  Raises:
      ValueError: if `axis` is neither -1 nor one of the axes of `output`.

  Example:
  ```python:
  import tensorflow as tf
  from tensorflow.keras import backend as K
  a = tf.constant([1., 0., 0., 0., 1., 0., 0., 0., 1.], shape=[3,3])
  print("a: ", a)
  b = tf.constant([.9, .05, .05, .5, .89, .6, .05, .01, .94], shape=[3,3])
  print("b: ", b)
  loss = K.categorical_crossentropy(a, b)
  print('Loss: ', loss) #Loss: tf.Tensor([0.10536055 0.8046684  0.06187541], shape=(3,), dtype=float32)
  loss = K.categorical_crossentropy(a, a)
  print('Loss: ', loss) #Loss:  tf.Tensor([1.1920929e-07 1.1920929e-07 1.1920929e-07], shape=(3,), dtype=float32)
  ```
  """
  if not from_logits:
    if (isinstance(output, (ops.EagerTensor, variables_module.Variable)) or
        output.op.type != 'Softmax'):
      # scale preds so that the class probas of each sample sum to 1
      output = output / math_ops.reduce_sum(output, axis, True)
      # Compute cross entropy from probabilities.
      epsilon_ = _constant_to_tensor(epsilon(), output.dtype.base_dtype)
      output = clip_ops.clip_by_value(output, epsilon_, 1. - epsilon_)
      return -math_ops.reduce_sum(target * math_ops.log(output), axis)
    else:
      # When softmax activation function is used for output operation, we
      # use logits from the softmax function directly to compute loss in order
      # to prevent collapsing zero when training.
      # See b/117284466
      assert len(output.op.inputs) == 1
      output = output.op.inputs[0]
  return nn.softmax_cross_entropy_with_logits_v2(
      labels=target, logits=output, axis=axis)


@keras_export('keras.backend.sparse_categorical_crossentropy')